|
|
||||||||
a Department of Pediatric Cardiac Surgery "Marta e Milagros," Azienda di Rilievo Nazionale e di Alta Specializzazione, Ospedale Civico, Palermo, Italy
b Department of Cardiac Surgery, University of Palermo, Palermo, Italy
c Department of Pediatric Cardiac Surgery, Ospedale S. Vincenzo, Taormina (ME), Italy
d Department of Pediatric Cardiac Surgery, Istituto Gaslini, Ospedale Pediatrico, Istituto di Ricovero e Cura a Carattere Scientifico, Genova, Italy
Accepted for publication January 2, 2008.
* Address correspondence to Dr Calvaruso, A.R.N.A.S. Ospedale Civico, Palermo Pediatric Cardiac Surgery, Piazza Leotta, 4, Palermo, 90127, Italy (Email: davidecalvaruso{at}hotmail.com).
Presented at the Forty-second Annual Meeting of The Society of Thoracic Surgeons, Chicago, IL, Jan 30–Feb 1, 2006.
| Abstract |
|---|
|
|
|---|
Methods: A series of 246 patients, from three joint institutions, having univentricular heart with restricted but not critical pulmonary blood flow received a bidirectional cavopulmonary shunt with additional forward pulmonary blood flow. All patients have been studied according to their progression, or not, to Fontan operation. Two hundred and eight (84.5%) patients underwent bidirectional cavopulmonary anastomosis as primary palliation. Twenty patients (8.1%) with previous pulmonary artery banding were also enrolled in the study. Patients who had received additional pulmonary blood flow through a previous systemic to pulmonary artery shunt for the critical pulmonary blood flow were excluded.
Results: No in-hospital death occurred. Follow-up was complete at 100%. Mean follow-up was 4.2 ± 2.8 years (range, 6 months to 7 years). During the observational period 73 (29.7%) patients, considered optimal candidates, underwent Fontan completion for increasing cyanosis and (or) hematocrit and (or) fatigue with exertion. Three patients expired after total cavopulmonary connection (3 of 73; 4.1% mortality rate). The remaining 173 (70.3%) patients are alive with initial palliation. All patients were still well palliated with an arterial oxygen saturation at rest about 90%.
Conclusions: According to our experience and results, bidirectional Glenn with antegrade pulmonary blood flow may be an excellent temporary palliation prior to a Fontan operation, which can be performed at the onset of symptoms. Bidirectional Glenn may also be the best possible palliation for a suboptimal candidate for Fontan.
| Introduction |
|---|
|
|
|---|
The bidirectional Glenn, alternatively named hemi-Fontan procedure, bidirectional cavopulmonary shunt, or bidirectional cavopulmonary anastomosis (BCPA) is a basic step toward total cavopulmonary connection (TCPC), preserving the confluence of pulmonary arteries, obtaining bilateral pulmonary blood flow, and leaving options for any subsequent procedures.
Presently, controversial data are available on the use of BCPA with antegrade pulmonary blood flow (APBF). Reported advantages of BCPA with APBF are a higher oxygen saturation, lower mortality, favorable effects on cardiac function, prevention of the arteriovenous fistulas, and a better growth of pulmonary arteries if compared with BCPA without APBF [1–5]. Previous studies demonstrated more frequent complications in the early outcome of patients with an APBF, but not to entail longer hospital stay [3, 6–8].
In 1991, Kobayashi and colleagues [9] reported the use of BCPA with an additional pulmonary blood flow in children who were not suitable for the Fontan procedure. Others suggested BCPA with APBF as a definitive palliation [4, 10].
The long-term role of additional pulmonary blood flow in the presence of a BCPA has not yet been fully studied. The aim of this study was to define the role of BCPA with APBF, through a banded or stenotic main pulmonary artery or through a ventricular-pulmonary artery conduit, in the surgical history of univentricular heart (UVH) with restricted pulmonary flow.
| Patients and Methods |
|---|
|
|
|---|
Patients
From 1998 to 2005, a series of 246 patients with single ventricle anatomy and (or) physiology receiving a BCPA with APBF were studied with respect to their progression, or not, to TCPC. Patients diagnoses are summarized in Table 1.
|
Hence, patients who underwent modified Blalock-Taussig shunt, or other types of systemic to pulmonary artery shunts, were excluded from our study, considering these kinds of shunt a primary palliation for critical pulmonary blood flow and a source of retrograde (and not antegrade) additional pulmonary blood flow, increasing total (and not effective) pulmonary blood flow. No patients with hypoplastic left heart syndrome were enrolled in this study because Damus-Kaye-Stansel-Norwood procedures are usually combined with a systemic to pulmonary artery shunt and we had no data concerning survivor patients with the Sano shunt in our series. Seven patients with Ebstein anomaly underwent successful BCPA.
Twenty patients received pulmonary artery banding before BCPA. Previous palliative procedures are shown in Table 2. In 208 (84.5%) patients, the BCPA with APBF was the primary palliation.
|
After division of the superior vena cava and transverse incision of the right pulmonary artery, continuous 6-0 polydioxanone running sutures (Ethicon, Somerville, NJ) for the posterior anastomosis and interrupted 6-0 polydioxanone sutures for the anterior anastomosis were used. The main pulmonary artery was left open. The azygos vein was ligated in all patients, excepting nine with azygos continuation of an interrupted inferior vena cava and moderate pulmonary stenosis. After weaning from cardiopulmonary bypass the pressure was measured directly in the superior vena cava (SVC) with the patient on 50% inspired oxygen fraction, to check Glenn physiology.
In all cases of SVC syndrome, the azygos vein was reopened to allow decompression of the superior cavopulmonary shunt into the lower body, avoiding the need of taking down the BCPA. We did not report any complication or significative change of arterial oxygen saturation after azygos reopening.
In three cases, pulmonary valve competence was reestablished with a valved connection for increasing cyanosis after BCPA due to a systemic recirculatory shunt through pulmonary valve incompetence, severe tricuspid valve insufficiency, and an atrial septal defect (Fig 1).
|
Postoperative Intensive Care Management
All patients were mildly sedated in the intensive care unit with midazolam and fentanyl citrate at 2 µg · kg–1 · min–1, and most of them received inotropic agents (dopamine and dobutamine at 6 µg · kg–1 · min–1) in the immediate postoperative period. Nitric oxide was used when mean pulmonary artery pressure is more than 18 mm Hg.
We considered mandatory the use of anticoagulant therapy since the early postoperative period (heparin intravenously 1 mg · kg–1 · day–1) and during hospital stay, continuing for the following 6 to 12 months postdischarge from hospital with aspirin (5 to 10 mg · kg–1 · day–1).
Pressure Monitoring
Pulmonary artery pressure, atrial pressure, and systemic arterial pressure were monitored continuously in the postoperative period. The SVC pressure monitoring is recommended as a marker of pulmonary pressure and pulsatile flow. To decrease the risk of thrombosis we preferred to avoid a central line in SVC for intravenous therapy, using only a single venous line for pressure measurement. In children weighing less than 10 kg, we used a surgical transparietal catheter inserted directly to the atrium after weaning from cardiopulmonary bypass. Pulmonary artery catheters were removed 24 hours after surgery without any complication.
Follow-Up Data
Patient records were reviewed and cross-sectional follow-up was carried out by direct review of patients data, and periodical hospital control by our physicians at 3, 6, and 12 months and then every year postoperatively. Follow-up visits included chest radiography, systemic arterial oxygen saturation measured by pulse oxymetry on room air, and echocardiography study. Follow-up was completed in all patients. Mean follow-up was 4.2 ± 2.8 years (range, 6 months to 7 years).
Statistical Analysis
Statistical analysis was performed using Statistica for Windows V.6.0 (StatSoft, Inc, Tulsa, OK). Continuous variables were reported as mean ± standard deviations (SD) or as medians and ranges, and were analyzed using the unpaired t test. Statistical significance was defined as a p less than 0.05. Percentages were given where appropriate. Linear regression analysis technique was used to evaluate correlation between variables. Kaplan-Meier methods were used for freedom from Fontan operation and actuarial survival analysis.
| Results |
|---|
|
|
|---|
The following complications were observed: prolonged pleural effusions (defined as drainage for more than ten days or the need for multiple drainage procedures) in 22 patients (8.9%); SVC syndrome (defined as the clinical manifestation of an obstructed SVC impeding the normal blood drainage from head, neck, and arms) in 24 patients (9.7%); pericardial effusions in 30 patients (12.1%); bleeding (defined as increasing rather than decreasing volume of chest tube output after four hours from operation) in 26 patients (10.5%); phrenic nerve injury in 7 patients (2.8%); and mediastinitis in 4 patients (1.6%). However, these complications did not require longer intensive care unit or hospital stay (Table 3). Mean arterial oxygen saturation in room air at discharge was 89 ± 3.7% with a hematocrit level of 42.8 ± 8.9%.
|
|
|
|
|
|
| Comment |
|---|
|
|
|---|
Over the years, the experience with the Fontan operation has increased [17–19] and has opened the way to a much wider application for its principle [20]. A number of major risks have been identified and managed by several modifications of the original Fontan procedure.
Correct timing of surgery in such patients is determinant of early and late outcome. No matter which type of UVH, the best staging, in our policy, is to keep the patient in their natural history as long as the systodiastolic function of the ventricle is preserved.
In UVH with favorable streaming, pulmonary to systemic blood flow ratio (QP/QS) about 1:1, and mild cyanosis, natural history might be preferable to surgical history. If any surgical palliation should be required before, a QP/QS of approximately 1:1 [21] should be obtained. In the presence of high pulmonary flow we prefer banding the pulmonary artery with the goal of maintaining the mean pulmonary artery pressure at less than 15 mm Hg; in case of low pulmonary flow, it will be increased by enlarging the narrowed native pulmonary outflow tract or interposing a valved right ventricle-to-pulmonary artery (RV-PA) conduit, to provide additional APBF.
Our goal is to try to achieve an effective-total pulmonary blood flow ratio of about 1. The more the effective and the total pulmonary blood flows coincide, the best the palliation. All the efforts are made to obtain the maximum oxygen saturation with the minimum ventricular overload.
In our study we analyzed the only sources of effective APBF available through a native, even if stenotic, pulmonary valve (n = 219), a banded main pulmonary artery (n = 20), or RV-PA conduits (n = 7; valved Dacron graft or bovine jugular vein from 12 to 16 mm diameter size). The exclusion of all other sources of pulmonary blood flow at the time, and before BCPA, is aimed to preserve the single ventricle, reducing the degree of atrioventricular regurgitation [21]. The advantage of the BCPA over the Blalock-Taussig shunt is to allow an increase of effective pulmonary blood flow without an increase in total pulmonary blood flow and cardiac work [22]. For all these reasons we consider BCPA with APBF a better step toward Fontan than BCPA with an additional pulmonary blood flow through a Blalock-Taussig shunt.
In our series, the presence of an antegrade source of pulmonary blood flow, allowed us to wait and perform the BCPA later, in older patients, as alternative to the Blalock-Taussig shunt.
Fontan operation has been extensively applied in one ventricle anatomy and (or) physiology and restrictive pulmonary blood flow with excellent results when patients were considered ideal candidates [23, 24].
In order to obtain the lowest mortality and morbidity incidence and the best long-term outcome, the ideal candidates to Fontan operation [14] should satisfy the following conditions: (1) a normal lung compliance (low pulmonary artery pressure, low pulmonary vascular resistances, absence of distortions of pulmonary arteries, absence of anomalous pulmonary venous connections and of restrictive or obstructive lung disease) [25]; (2) a normal heart compliance (without ventricular hypertrophy related to volume or pressure overload and subendocardial ischemic damage) [26–30].
Patients are considered less than ideal for the Fontan operation in case of insufficiency of the systemic or common atrioventricular valve, presence of congenital or iatrogenic abnormalities of proximal or distal pulmonary arteries, elevated pulmonary artery resistance (>2 Woods units) and (or) increased pulmonary artery pressure (mean >15 to 18 mm Hg), low ejection fraction, and multiple major aortopulmonary collateral arteries (Table 4). A staged surgical option for these patients should be the BCPA with APBF because it allows an adequate growth of pulmonary arteries and veins [31] thus preparing optimal hemodynamic conditions for the Fontan procedure [32].
Our policy is not to perform BCPA early in infancy to preserve, until possible, the natural history of patients with UVH and a source of APBF. Older patients benefit more in terms of durability of the palliation than in terms of enhancing the systemic oxygen saturation from BCPA, for the disadvantageous upper body to lower body ratio.
An APBF source, if included in the pulmonary circuit, might be also able to produce some increase in cardiac output under exertion and carries the proper liver metabolites, which maintains normal pulmonary arterovenous continuity, anatomy, and function [21, 33].
Seventy-one palliated patients of 173 waiting for Fontan (28.8% of the whole series of 246 patients), have been considered less than ideal candidates for TCPC. So the question for them should be, Is the perfect the enemy of the good? That is, whether patients with BCPA and APBF must necessarily be converted to Fontan circulation or not. In our opinion, patients at high risk for death or poor outcome, because less than ideal candidates for TCPC, must be excluded for Fontan completion. Ideally, a more appropriate selection of patients will result in a more favorable impact on early- and long-term outcome.
It is true that the arterial oxygen saturation cannot reach normal levels with BCPA and APBC, but considering the zero mortality after surgery, the good status in our series patients, an oxygen saturation in the 90% range, the stability over the time of the palliation, and the age distribution of the patients at the last follow up (Fig 5), we could consider this palliation as definitive for high risk patients for Fontan.
In conclusion, despite that we considered extracardiac Fontan the technique of choice for patients with functional single ventricle [34, 35], not all patients with UVH should undergo Fontan completion. Zero surgical mortality should be considered mandatory in UVH with normal pulmonary arteries and absent cardiomegaly, after appropriated and correct staging [36].
In patients considered suboptimal candidates for a Fontan procedure but in whom the Fontan criteria might be met in the future, BCPA and APBF should be considered as the first stage palliation. If for any reason the Fontan is considered a high risk procedure, BCPA with APBF should be considered the best possible palliation. Further long-term studies are certainly needed to clarify whether BCPA and APBF might be considered, or not, a definitive palliation for a suboptimal candidate to Fontan.
| Discussion |
|---|
|
|
|---|
DR MARCELLETTI: We had to do an urgent resternotomy in three cases while removing the atrial lines.
DR BACKER: So three patients out of how many?
DR MARCELLETTI: Three out of 246.
DR BACKER: Let me survey the audience while we are discussing this. How many people are still using right atrial lines, direct central lines into the atrium that are then pulled out late after the operation? (Show of hands.) It looks like at least 60% of the audience. And how many people are avoiding the use of right atrial lines? (Show of hands.) Just a select few who probably have had complications.
DR C. KNOTT CRAIG (Oklahoma City, OK): First of all, congratulations. You continue to impress us with your results and the evolution of the Fontan operation. I noticed that there was a fairly high percentage of patients with SVC [superior vena cava] syndrome. How did you deal with that, and was it necessary to limit the pulmonary blood flow in those patients?
DR MARCELLETTI: Well, we did not remodulate pulmonary blood flow after the appearance of superior vena cava syndrome. But what we have done, we have gone back and opened the azygos vein in some patients. I think that the question you posed to our colleagues previously in what to do with the azygos vein is a very relevant question. I think that the azygos is a very good safety valve. Also, in small children, when you do a Glenn under four months of age, I do not ligate the azygos vein. I think that its very safe.
And the superior vena cava syndrome came mostly when you had a pulsatile pressure in the superior vena cava, lets say, of 28, 30 over 15. And then if you just release the azygos vein, everything came out pretty well.
DR BOHDAN MARUSZEWSKI (Warsaw, Poland): Carlo, I still remember that you were the one who taught me and trained me how to do extracardiac Fontan quite early. But I think one comment to the line left in the right atrium. We could not leave the line if this is a systemic atrium; so that I believe should be avoided. But otherwise we leave all the lines, especially in neonates. Now, in the late 90s, if you let me have some comment, we were following the advice from quite eminent cardiologists in Europe, and we were considering some patients who had very successful Glenn anastomosis with additional shunt, forward flow through the shunt, or banded patients, as the final palliation. Now, these patients started coming after six, seven years, coming back, desaturated, and now we finalized the TCPC [total cavopulmonary connection] and we do the extracardiac Fontans in them, in those who are still operable. My finding is that these patients are becoming bad candidates for Fontan because the pulmonary arteries, they do not grow. And we looked at the catheters when they were performed six years ago and now, and actually those PAs [pulmonary arteries] remain the same and the patients are much, much bigger. So, of course, we have to fenestrate all of them. And in our early Fontans, like 10, 12 kg, we very rarely had to fenestrate. So what do you think about the final palliation in the sense of the natural history of these patients?
DR MARCELLETTI: I thank you very much, Bohdan, and I came back to a slide, which I like very much. I think that patients born with one-ventricle physiology or anatomy, having a normal arborization of the pulmonary arteries, should be offered zero mortality for the Fontan. If, by any chance, the mortality that you can offer is over zero, you have to think about that very thoroughly, because we have seen beautiful postoperative courses in patients who have these characteristics. I like to remind this, because I think its a part of my personal philosophy. Patients should have normal pulmonary arteries and they should not have cardiomegaly. I think that when a child comes to us with the palliation being bidirectional Glenn and a contralateral shunt, with a big heart, they have a very high risk. And then you have to work on the appropriate staging; that is, how long do you let the patient stay with a Glenn and antegrade pulmonary blood flow? Again, I think we have to go back to very simple consideration; hematocrit and cyanosis. And the heart must be small. What we have done in the past, making a mistake, was to do Fontans for everybody. And I think thats not really the best approach. Also, you know very well, Bohdan, that I do not fenestrate any Fontan, because I think that if you have to fenestrate you are dealing with a suboptimal candidate, and maybe you would be better with a Glenn and maybe a small tube from the ventricle into the pulmonary arteries, which is an excellent palliation.
DR MARUSZEWSKI: I just want to add one comment. What worries me a little bit is the initial report of the Fontan multi-institutional study that was given by Steve Colin, where, in patients who have just a Glenn and the forward flow, you may expect 150% to 200% of the volume load on the ventricle. So I think we should keep in mind that we are overloading these ventricles. Of course, if they are not candidate for Fontan, this is the compromise.
DR MARCELLETTI: I think that since we have adopted strictly this philosophy, we are not reporting 100 consecutive Fontans with no failure and all patients in very good condition.
DR THOMAS YEH (Dallas, TX): Congratulations on your study of this provocative issue. We were similarly concerned when we assessed 435 patients undergoing cavopulmonary shunting in Toronto, and found that survival was essentially equivalent between patients who underwent cavopulmonary shunt but could not progress to Fontan and those who underwent cavopulmonary shunt and were able to progress to a Fontan. Our presumption was that "high risk" patients were the ones who did not progress to a Fontan. Nevertheless, this high risk subgroup had essentially equivalent survival to the "better risk" patients that did go forward for the Fontan procedure. This led us to question, as you are, whether or not Fontan completion is always warranted.
DR MARCELLETTI: Agreed. Thank you.
DR MRINALENDU DAS (Kolkata, India): Do you differentiate between right ventricle and morphologic univentricular heart and left ventricle morphology? Do you consider the right ventricle morphology the worse candidate and you dont complete the Fontan in those cases?
DR MARCELLETTI: No, basically, we dont take into consideration being a dominant right or a dominant left ventricle. I think that one principle is that when there is a small pulmonary ventricle gathered by a tricuspid valve, and if it is small, should never be sacrificed. Thats another point. I think that even a ventricle with a tiny tricuspid valve should be inserted into the circulation because it keeps bringing metabolites from the liver, keeps stability of the micropulmonary circulation, and probably it does provide some increasing cardiac output.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. P. Graham Jr The Year in Congenital Heart Disease J. Am. Coll. Cardiol., October 28, 2008; 52(18): 1492 - 1499. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |